PROJECT SUMMARY/ABSTRACT Increasing attention is being paid to the size of provider networks in private health insurance plans. The health insurance exchanges brought about by the Affordable Care Act (ACA) are a proximate cause of the shift by insurers to scrutinize provider network size and composition, but greater scrutiny of health care costs by companies has also resulted in a ?rediscovery? of provider networks as a potential means of cost control. We propose to take advantage of the roll-out of a set of network-based products by a large national health insurance carrier in the large-employer market. The roll-out involves a geographically diverse set of states and an array of insurance products. The products provide variation along several important dimensions: 1) network size: in each market there will be differences in the size of the selectively contracted network relative to the traditional network providers; 2) primary care provider (PCP): some plans will require that the member choose a primary care provider; 3) out-of-network care: depending on the product, out-of-network providers will be accessible at lower benefit levels or will not be covered; 4) specialist referral: some products require that the member obtain a referral in order to have coverage of specialist care. We specifically seek to address what the effects are of narrower network designs on cost, utilization, quality, and access to care. Sub-aims will examine how effects of narrow network designs affect care for those enrollees with chronic conditions. Detailed data will be obtained from the health insurer including but not limited to: enrollment information, demographic information, plan details (deductible, out-of-pocket maximum, coinsurance, etc.), provider network information, and health care claims data. Our data will feature a large number of companies two years prior to the introduction of the new products (January 2013) through two years after the introduction of the new products (December 2016). With these data we will compare the experience with the new products to a set of employers who stayed with traditional network plans during the same period. Our proposed research will provide timely and comprehensive information on a topic of high policy relevance.